Tarmac SL9 | Engineered to Win

The Tarmac SL8 is already the winningest bike in the WorldTour. So how do you make it faster?
In this episode, guest host Ben Edwards sits down with members of the Specialized engineering and product team at the global press launch for the new S-Works Tarmac SL9 on the Costa Brava.

Ben is joined by Lio Bardina, aero engineer; Denis Kürner, senior design engineer for road performance bikes and lead engineer on Tarmac; and Alex Jerome, category leader for road and gravel products.

Together, they go deep inside the development of the new S-Works Tarmac SL9, a bike engineered to be the fastest to the finish in the more aggressive road racing of today and the future.

The team explains why the goal was not simply to make the lightest bike or the most aero bike, but the fastest bike to the finish line. That meant using the equation of speed to optimize aerodynamics, weight, rolling resistance, ride quality, handling, stiffness, and real-world rider data to deliver the fastest time to finish and the Tarmac promise: ‘One Bike to Rule Them All’.

They discuss how racing has changed since the launch of the Tarmac SL8, with higher speeds, longer breakaways, and decisive race moments happening farther from the finish. Those changes shaped the design of the SL9, including its narrower Speed Sniffer head tube, reshaped Flow Fork with deeper, inwards twisted fork blades, dropped downtube, S-Works Rapide aero seatpost, Win Fin, and refined tube shapes.
The team explains why the Tarmac doesn’t need to look like a trending aero bike to be the fastest bike in the world. Rather than designing for appearance or optimizing for the fastest bike without a rider on it, Specialized modeled what actually determines time to finish: WorldTour courses, rider power files, real wind angles, rolling resistance, and data from the velodrome, road, and the WinTunnel, including testing with a moving-leg mannequin.

One example makes the impact clear: based on simulation of Demi Vollering’s 2024 Tour de France Femmes performance, the SL9 could have made her 14 seconds faster over the final 83 kilometers of the race. In a Tour decided by four seconds, that is the difference between losing and winning yellow.

The episode also goes inside the engineering challenge of improving aerodynamics while preserving the ride quality athletes already loved on the SL8. The S-Works SL9 delivers four watts of aero savings at 45 kilometers per hour, while the frame gains just two grams, coming in at 687 grams.

This is the story of how Specialized updated an icon: not by chasing a single number, but by building the fastest Tarmac ever for the moments that decide the biggest races in the world.

This episode covers:

• How modern road racing has become faster and more aggressive
  
  
• Why long-range attacks and solo breakaways influenced the SL9
  
  
• Why the goal was fastest to the finish, not simply lightest or most aero
  
  
• How the equation of speed shaped the development process
  
  
• Why real-world wind angles matter more than traditional aero assumptions
  
  
• How Specialized uses rider power files and race data in simulation
  
  
• Why the SL9 does not need to look like a traditional aero bike
  
  
• How the team achieved four watts of aero savings at 45 kilometers per hour
  
  
• How the new head tube, fork, seatpost, and Win Fin improve speed
  
  
• Why the moving-leg mannequin changed the way the bike was designed
  
  
• How athlete feedback shaped ride quality, handling, stiffness, and compliance
  
  
• What the SL9 simulation showed from Demi Vollering’s 2024 Tour de France Femmes
  
  
• Why small gains can decide the biggest races in the world
  
  

Guests:
Lio Bardina
Denis Kürner
Alex Jerome

Subscribe to Specialized Bicycles on YouTube for more conversations with the engineers, athletes, and creators shaping the future of cycling.

Get More from Specialized

TikTok: @specializedbicycles

Instagram: @iamspecialized

Specialized MTB Instagram @iamspecialized_mtb

Specialized Factory Racing Instagram: @specialized_factoryxc

Facebook: specializedbicycles

Outride: @outride

Transcript

SPEAKER_01 0:03

Welcome to the Specialized Bicycles Podcast. I'm Ben Edwards. I'm filling in for Ben Caper, and that's because we're here on the Costa Brava at the launch event for the Tarmac SL Mine. That's right, there is a new tarmac, and I'm sitting here with much of the engineering team and product team that helped put the tarmac SL Mine together. So we're going to go deep into tarmac. Hey guys, why don't you introduce yourselves for our listeners though? So I am Ulio Berdina.

SPEAKER_03 0:27

I am the aero engineer here at Specialized.

SPEAKER_00 0:30

Dennis Kurner, senior design engineer for road performance bikes and lead engineer on Tarmac SL9.

SPEAKER_02 0:36

I'm Alex Strome and uh the category leader for road and gravel product.

SPEAKER_01 0:41

All right, so so right now it's May, and like I said, we're at the global press launch here on the Costa Bravo for the bike, but you'll be listening to this right before the Tour de France starts, when we've really revealed the tarmac SL Mine to the world. And so what I really want to do is after we've introduced this crew, we've established where we are, I think I want to talk a little bit about tarmac. Not the new one, but tarmac as a this iconic bike and what it's been to the specialized brand. So tarmac obviously it's the winningest bike on the world tour since it launched. It's won more world championships than any other road bike. When you're tasked with updating a tarmac, it's a fairly big deal for specialized, it's a big deal for our athletes, it's our pinnacle race bike. What are the feelings when it's time to bring a new tarmac to life?

SPEAKER_00 1:25

I've been working in the bike industry for a little over 13 years now, and tarmac is always, I think, um, kind of the benchmark when it comes to a like um world tour race platform. So honestly, like you feel honored, but you already know that it's going to be a huge challenge that you're going to be facing for the next three years. So it's mixed feelings, excitement, but also a little bit uh scared almost.

SPEAKER_01 1:50

So you're right, it is this legendary bike that Demi Vollering, Remco, you know, Lotta, Lorena, the top riders in the world are really relying on all of you to give them that advantage. So let's talk a little bit about the kind of racing that we're seeing that those athletes are up against. Because you know, when Tarmac SL8 launched, you know, we we've seen the the pro Peloton evolve since then. You know, what are some of the things that you've seen out there in pro racing since SL8 was developed that you know influenced how you created SL9?

SPEAKER_02 2:22

Probably the most critical piece for us is just in terms of how racing has evolved, just the speeds continue to get faster and faster. We see uh obviously tour stages, one-day races, they just um average speeds continue to get quicker. Key moments within the race continue to get quicker. We see obviously, let's say, more aggressive racing at some points with key athletes also. So developing a platform around that and making sure that it's uh yeah, focused on those kind of race-defining moments and key moments with within a race is a pretty key priority for the team.

SPEAKER_01 2:57

And it seems like maybe this is something for you, Leo, as an aerodynamicist. We're seeing riders breaking away earlier, it seems to me. They're spending more time alone than they did in the past.

SPEAKER_03 3:08

Yeah, of course. And that's a key differentiator, right, of the new chassis, looking back to what tarmac used to be is we now talking with S-Racing, working with our athletes, with our pro riders, realize that in modern racing, they're doing these really long breakaways from pretty far out. Anytime a rider is breaking away, they're not riding with two bottles on their bike, they're tossing that back bottle and relying on the team car quite a bit. So that's where the wind fin comes in on the SL9. And really, that was a major priority for us is developing this chassis for exactly how our athletes use it.

SPEAKER_01 3:40

So you're you're going deep onto some technical detail right there that we'll we'll get into later for sure. But that's the kind of thing, right? You're actually literally when you make this bike. We say maiden racing around here, it's it's not really a slogan, it's the fact that you are all watching the races every day and you see how the racing, the the new tactics that are being used and the strategies the riders are employing, and that's what you're thinking about when you're thinking about how to evolve this bike. Tarmac SL8, pretty special bike, pretty successful bike, you know, world tours, grand tours, monuments. You know, it's one just about everything there is to win. What made that bike so special for SL8? And then the second question after that is how do you actually upgrade, improve the performance on a bike that is already at such a high level?

SPEAKER_02 4:24

I think for the SL8, obviously it's a fast bike. That's uh, you know, something we need to have in those platforms. But a couple other key pieces on SL8 that I think we all um are pretty proud of and feel like it does quite well is obviously a weight perspective. It's a really light chassis. It's probably in terms of race bikes in the world tour, it builds up to the lightest platform. And then ride quality is really a key piece for us and something that we get a lot of feedback from our athletes and world tour teams on. So that's something we wanted to make sure that we retained and brought into SL9 was not only the stiffness that SL8 had and kind of those handling characteristics, but also the compliance aspect, especially over those one-day races and then certainly over grand tours, making sure that that bike has the same amount of comfort and compliance within a race platform was a key piece for us for SL9.

SPEAKER_01 5:13

So I I think it's safe to say, right, with tarmac SL8 and SL7 before it, it was really that one bike to rule them all concept, right? No matter what the day looks like, you're trying to win a monument, you're trying to win a grand tour, you're on the right bike for that day, right? No compromises, not making a selection between a light bike or an aerobike, you have it all built in. So when it comes to SL9, safe to say that's still the philosophy. It's still a one bike to rule them all philosophy.

SPEAKER_00 5:38

Yeah, I mean, definitely. Yeah, uh, I think that's the philosophy that we all um agreed on that we want to focus for uh the next generation of tarmac. And um in the beginning of the project, I think the goals were clear that we are already able, or the proteins are already able, to build the SL8 at the uh UCI weight limit of 6.8 kg. So it was not a goal to remove weight or take weight out of the frame. But uh with the um speeds that we are we've seen in road racing in the recent years, it's just continue to get faster and faster every year. Um I think it was a clear goal that we have to uh improve the aerodynamic performance of the of the next generation of tarmac.

SPEAKER_02 6:20

Yeah, we get a ton of help from our simulation team in terms of kind of defining where our goals should be in terms of trying to help the riders get on the fastest platform and also basically have the fastest time to finish within those different courses or segment of courses. Um that's a key piece for us. And actually I think when we went through that process, obviously to Dennis's point, the goal was to make the bike faster and certainly add from an aerodynamic perspective. But um we were willing to basically have what, a 50 gram weight budget to make that happen. So that was kind of the key piece that we got from the data of time to finish was basically pull out as much aerodynamic as we could from that chassis, but with a 50 gram weight budget.

SPEAKER_01 7:03

Yes, and that's something I really want to dig into, right? This idea of equation of speed and time to finish and really the tools you're using to develop the bikes, because you look out there at the competitors, right? And just to say it, when you look at the the podium of the Tour de France, there's you know two guys that are riding full aero platforms, you know, and they've whatever the bikes weigh is what the bikes weigh, and they've they've chosen to go full aero. You've obviously made a different choice to create what you believe is the fastest bike in the world. And you know, is it is it correct to say that when you go into this and you're looking at tarmac SL9, you're not trying to make the lightest bike, you're not trying to make the most aero bike, you're not trying to make a bike that does well in some lab test on the test bench. You're just trying to make the fastest bike. Elaborate on that a little bit before we leo. I think that's you know, that's a really unique distinction because we see so many people looking at a wind tunnel test. Yeah. And they say, oh, it won the wind tunnel test, and they actually substitute the word fast for arrow, right? So something's arrow, that's gotta be the fastest, right?

SPEAKER_03 8:04

Yeah, I mean, that is something we work very in-depth on, right? We with SL9 have full world tour course simulations, we have critical moment and race simulations, we're going through all these processes to see what gets us across the finish line the fastest, right? And that's where we say tarmac wasn't built to be an aero bike, tarmac was built to be a fast bike. But I think an important distinction to keep in mind there is just because tarmac may not look as aggressive aero as other race bikes out there, we have arrow benchmarked it against those fastest, most aggressive style. Even though tarmac may not have that most aero-aggressive styling, doesn't mean it isn't one of the most aero bikes on the market. We actually benchmark probably the three most aggressively styled aero bikes on the market where as a consumer you'd look at and say, I think this is the most arrow just based on where it looks. And we beat two of them and were within a hair of the first, right? So we are playing right there in the same ballpark as the three most arrow-looking bikes on the planet with a bike that's like half a kilo lighter. So if we're just as arrow and we're a lot lighter, we're gonna be a heck of a lot faster.

SPEAKER_01 9:11

Yeah, interesting. So so the fact that maybe it doesn't have those deep airfoils, the fact that it doesn't look as outlandish maybe as some of those bikes, when it comes to the real world and the yaw angles we see, and when you actually put a person on the bike, it's not actually the the visuals of a deep airfoil may not actually be the fastest way to get to the finish line.

SPEAKER_03 9:31

No, that's correct. So we can obviously, if we wanted to, it'd be really easy to make a tarmac that looks arrow, but to us being arrow is more important than looking arrow, right? So it's not just throwing more surface area at the tubes, making all of them deeper. Um, we talk a lot, right, about deeper in the front, lighter in the rear. We could add a ton of weight to our chainstays and seat stays and get basically no benefit out of them once there's a rider on the bike. So we're not gonna do that, right? And that's where the tarmac family really is unique in the market is we're adding arrow without adding weight. Where do we make our bike arrow? It's making the head tube probably one of the thinnest head tubes of any bikes in the world tour. Our fork is incredibly arrow without having that much surface through its twists and it through its unique shaping. So that's something that we have really pushed and really innovated on is adding arrow without adding weight.

SPEAKER_01 10:22

So it's arrow where it matters, not arrow where it it looks cool and it just ends up being in the dirty air of the rider and and and not providing any benefit. It's really that leading edge.

SPEAKER_03 10:31

No, absolutely.

SPEAKER_02 10:32

Yeah, but I think a key piece there is we're not just to Leo's point, we're not um creating the the bike via design. We're creating the bike via data. And one of the key pieces for us is the protocol in terms of how we actually do our wind tunnel testing. I think um maybe you're better suited to speak to that, but we try to use much more realistic real-world data that we've collected, um, our teams have collected, and that really helps us actually, I think, approach it from a bit more realistic standpoint.

SPEAKER_03 10:59

So I guess to elaborate on that, right? I came to cycling from automotive aero. I was an aero engineer at a semi-truck company, and I thought it was hilarious when I came to cycling. We in the cycling industry tended to use the same exact wind average weighing function as they use for semi-trucks. I'm not sure if you've ever seen a semi truck, and it is a lot taller than a cyclist. So that automotive standard used for semi-trucks that was just blindly adopted by the bike industry assumes your average height is three meters off the ground, right? We looked at that, dropped it down to one meter off the ground using the same methodology. And what is the impact of that, right? The impact is your average wind angles a rider will see are much lower. They're much more heavily weighed to zero, while for other companies they place less importance on zero and more at very high wind angles. If you optimize a bike for extremely high wind angles, those deep tubes have that opportunity to perform that sailing effect. But if you're optimizing more heavily towards lower yaw angles, really where almost all the performance comes out of is frontal area. And I'm sure as Dennis will speak to later, there are some really unique ways that we manage to pull frontal area out of this chassis without adding any weight.

SPEAKER_01 12:10

That's wild. So if you're using the automotive standard, you're assuming your bike is three meters off the ground.

SPEAKER_03 12:14

An average height of, yeah. I mean, look how tall a semi-truck is and look at how tall one of us is, right?

SPEAKER_01 12:20

And clearly those, and to also to be clear for people listening, just in and for myself as well, you're talking about that weighted wind average. That is the angle of wind that is going to be impacting a rider on the bike as they're riding a meter off the road.

SPEAKER_03 12:31

Yeah, absolutely. And I think a picture is worth a thousand words in this case. If someone wants to refer to our white paper, we have that wind average weighing charts of us compared to what's used in the industry and other publications. And it's really stark how different we are to a lot of them, but that is a data-driven approach. We validated this with different university studies. We've had wind sensors on team cars, um, we've done a lot of a lot of studies to make sure that our assumptions are something that we feel very comfortable with. Because at the end of the day, like we were talking about, what wind average you design a bike towards can have a huge impact on what you end up with. If you design a bike for Kona, where you do expect those very high average wind angles, those super deep tube sections actually do make a lot of sense. If you look back to the Shiv Tri, right, that's what Shiv Tri was designed for, and that informs that look. If we look towards tarmac, that is not at all what tarmac is developed for. Tarmac is developed towards average wind conditions expected in world tour races, and that's why tarmac looks the way it does.

SPEAKER_01 13:30

Yeah, interesting. So if we were using that same three-meter, you know, position that the rest of the industry is using, tarmac might look a lot different.

SPEAKER_03 13:39

Tarmac would absolutely look a lot different.

SPEAKER_01 13:41

But it would actually have less real-world application on the world tour.

SPEAKER_03 13:46

Yeah, and that's also doubly true as we see world tour speeds increasing, right? So if you look at a sprinter sprinting on a tarmac, they're gonna be going a lot faster than that average 45 kph. Yeah. The faster you go, the lower those average ones needs are. If you're doing a solo breakaway at over 50k, again, that lowers those average yaw angles as the rider speeds increase. So as we get into those more critical moments in a race, uh, we're weighing even more heavily to zero than our assumptions. We did a lot of that initial development around R, which is why in our final validation of the bike with those actual rider simulations using their power files, we see that this bike performs extremely well.

SPEAKER_01 14:25

Yeah, so actually I want to I want to kind of turn our attention to that. We we kind of went deep on arrow for a minute there, but it's actually really fascinating that idea that you know, three meters versus one meter off the ground and where we're actually riding our bikes. Um, but that's I want to eat this arrow is we talked about it's kind of one piece of the puzzle, right? And it's not the only piece we're solving for. There's weight, there is that ride quality, all those other pieces of it. So in and I think you guys have the equation of speed, right? That's we saw it with crux, you know, when crux came out, diving into the equation of speed and how it works within gravel. Could you explain a little for our listeners about equation of speed and and how this simulation works? What are the inputs for the simulation? Um, how important is it we get those inputs right? Just some background on that.

SPEAKER_03 15:07

Yeah, so the equation of speed, we can kind of bucket our inputs into a couple categories, right? You have your rider input, that's how heavy is your rider. When we have that rider in our wind tunnel, what's their CDA? And of course, what is their power profile? We get their fit files directly off their head unit that we can use from any race they've competed in. Um, your next bucket is your course bucket. What are the wind conditions of that course? Uh going back to the crux, we looked a lot at unbound where winds are actually very high on average, so we're gonna bring that into that course. Um, same thing for surface roughness of each course. That's a huge differentiator between road and gravel. And then we have a full 3D profile again, developed from their um head unit files. So we know the exact elevation and curvature of each course. Based on those, then it's the part that we control, it's the bike. That's what's your rolling resistance, what's your drag, uh, what is your weight, what are all the aspects of the bike that we can change ourselves? And then, based on all of those inputs, you can then calculate the time to finish. And more interestingly for us, we can vary those inputs. If Dennis has an idea to change the head to you and he does a quick investigation, he thinks, hey, maybe this adds 15 grams, so we can get one count out of it. Does that make you faster or slower? And really, when we talk about this being a data-driven bike and not a design-driven bike, that's what we mean. Every single decision made in the development of tarmac comes back to that question of if I do this, is the bike faster?

SPEAKER_01 16:32

So, talk to me a little bit about that, the the one aspect of it, the the route aspect. You talked about using unbound for crux, which makes perfect sense. Um, what is the route we're using for for tarmac SL Mine to create that bike that can go out and win a monument and win a grand tour?

SPEAKER_03 16:48

So there are kind of two different sets used. Um, in the development of the bike, chassis, um, Marcel Kaiser, a simulation engineer, developed a statistical average of all world tour and monument stages scaled to be 100 kilometers long, which sounds really fancy, but basically he's just kind of bringing in every single course, seeing how on average, what are gradients, what are curvatures, what are the features of the course, and scaling that down to reasonable length to have a time delta from. So that's how we develop the chassis, that's how we made the decisions along the way. But at the end of the day, when it's time to show the bike to our pro riders, show the bike to our consumers, and prove that it really is fast, it makes to us more sense to look at actual defining moments in races. So that is demi-breaking away solo one albus, that is um finish of Malign Sign Rainbow, that is actual finishes of races where riders are seeing wind, either they're riding solo in a breakaway, they're sprinting in those key moments using actual rider data. How would this bike perform relative to both the SL8 but also our competitors that factor one, the Cervello S5, and the Y1?

SPEAKER_01 17:55

You're using the equation of speed, and you know, I think about almost you know, this idea of being like a record producer, right? And you can, you know, you can dial up the arrow, and then if you dial up the arrow, but then the weight goes too heavy, all of a sudden the time to finish gets slower. So it's all about shifting those dials, right? So it's not an arbitrary, like we're gonna try and make it this much more arrow and we're gonna try and make it this much lighter. You know exactly how much more arrow and how much lighter you need to make it, is that right?

SPEAKER_03 18:19

Yeah, and a great example of that, right, is we change the spec in our tires. We're specing 30 mil cotton TLRs, where in the past we would have spec's smaller sizes of rapid air turbo TBR tires. Um, and that was based on sure, they're a little heavier, they're a little bit more drag, but we know the increase or the benefit in rolling resistance is going to outweigh any penalty and weight and arrow. So down to the tires, anything that we spec on this bike or any decision we make on this bike, is it faster?

SPEAKER_01 18:48

And and so that's why we actually see our World Tour athletes on the 30 mil tires.

SPEAKER_03 18:52

Absolutely. Even on the TTs, they're running the 30 mil tires.

SPEAKER_01 18:55

Interesting. So we we've talked about what can happen in in simulation. Uh obviously the riders are a huge part of this, you know, and in rider feedback. So maybe, Alex, can you talk a little bit about the feedback you get from the riders to make sure that it's the ride quality they're looking for, it's the geo they're looking for. What's what's that relationship like?

SPEAKER_02 19:15

Yeah, we are we're super lucky to have close connections to a lot of our riders and just all of our world tour teams. So it makes it really easy to gather feedback and just to stay connected. I think if from their perspective, they trust us to handle, let's say, the arrow aspect and run the simulations and have Leo and Marcel really kind of give direction to what the product needs from that perspective. But to your point, we're really talking to them about how they're getting fit on the bike. You know, we we fit every single one of our World Tour athletes in Morgan Hill at team camps in December. So we have a ton of data from that and we get a ton of feedback from them. We also have quite a few athletes that have been with us for numerous years. So they've been with us over a couple different generations of platforms, and it just allows us to get really nice hands-on feedback. And then with each new platform, we try to get a lot of those riders on it early as well, just to influence how the product turns out and just um yeah, making sure that the product's right for them. But from a geo perspective, we actually haven't really changed anything on SL9. We continue to have really nice success, both from an error perspective in the wind tunnel with Leo, but also with like Jason and Whalen, the individuals that fit a lot of our World Tour athletes and do a lot of the physiological testing. Um, yeah, we get great results with the fit that we have on those bikes. So we didn't change any of the geo. And then I think probably the third biggest priority for us within the project was retaining that kind of compliance and handling characteristics that we had with SL8. It's pretty much from all of our athletes and World Tour teams, they're they're quite happy with where we landed there. So one of the key pieces, even though we were changing basically every single tube shape on the platform, we wanted to make sure that we retained the same stiffness aspect through the different frame sizes, and then also specifically the compliance on the rear end as well.

SPEAKER_01 20:58

Aaron Powell So safe to say the athletes, you think like Remco and Demi, right? They've both been on SL6, SL7, SL8, and now SL9, they were pretty clear in terms of how compliant it felt, in terms of the way it felt to the pedals, in terms of the handling. They said, don't mess with it. We love it, it's perfect.

SPEAKER_02 21:15

Yeah, and that's you know, it's easy feedback from them, but we're also lucky um, you know, we have quite a few athletes that come from other programs as well that have ridden other chassis from different brands. So it's nice to get feedback from those folks as well when they start to ride on our bikes and make sure not only it's you know handling and feels good, um, hopefully better, but uh it's just nice to get kind of different points of view. But we obviously have our long-standing athletes that are really happy with kind of where we've landed from the handling perspective and certainly the compliance perspective.

SPEAKER_01 21:45

Well, so so so big buckets, right? We know we've we've we've got the the handling, the geo, the compliance, that ride feel that the riders want. You knew what those numbers were, had to stick with that. Um comes to the arrow, right? You knew you wanted to make. The thing faster based on what the equation speed was telling you at faster, more aerodynamic, not the same thing. But then weight. You couldn't add any weight, right? What was the equation of speed telling us about the weight of the bike? Because as you said, tarmec is already the you know the lightest of the all-around race bikes out there.

SPEAKER_00 22:19

We could have added a little bit of weight actually to achieve those four counts of um arrow improvement that we actually uh settled on at the beginning of the project. So we gave ourselves a 50 grams of weight um budget to add on the frame um for arrow features because we were already expecting, okay, we have to increase the surface area on the frame. We will most likely have to add some features like we see now. For example, the drop-down tube. Uh Winfin, we know, okay, this is going to be coming at a weight budget. So um, like what you said, uh tuning the record, basically, we can um tune the aerodynamics, reduce the aero drag by four counts, get the weight up by 50 grams, and um calculate the time to finish. So grams would have been uh still acceptable and we would have achieved our goals basically.

SPEAKER_01 23:08

You still would have had a faster bike, but when you look at the numbers, you didn't add 50 grams, you added two grams. Yeah, we had to add two. Very unfortunate. So let's talk about how that happened. I mean, you're going into a project where you're saying there's no way I can deliver the aero benefits you want without 50 grams. Yeah. Give me some extra weight. I need some extra material. Yeah, there's more surface error in this bike than on the previous bike. Yeah. So how did you do it?

SPEAKER_00 23:33

Uh, you mean how we came to the to the 687 gram? Yeah. Um, I think that's probably everything has to do with maybe we elaborate on flow state design a little bit later uh during this discussion. Um, but we developed a pretty nice process, I would say, starting with ASOS 1. Um over tarmac SL8, ASOS 2, we just keep on evolving this process. And we have a pretty strong simulation team. Um, basically, for tarmac SL9 is what they um can do is that they have a ply-by-ply analysis and FEA combined basically in a simulation environment. So they can really um tweak the layup, we define the shape of the bike so that um every single ply is basically serving double duty. So it's not just stiffness plies or structural plies, but every single ply is basically serving um both uh properties. So um we went to uh the preforming room to the factory basically. We spent a full week with the workers in the preforming room um when they built the first frames basically, and then the frames go into the mold. Two days later we have the first um frame on the test bench, and that frame came in at 710 grams actually. So that's um 25 grams already below our maximum frame weight that we gave ourselves. And we put it on the on the test bench, and it was hitting the stiffness targets from the simulation forecast by plus minus 1%. So that was already a pretty uh special moment in the project, I would say. So that was also the moment where um we as a team came together and said, okay, we are at 710, let's just try one more time, uh, one more push basically, challenge the simulation team and and see what else they can what they can do. And they came up with a with a um new layup basically and had the 687 gram frame weight as a forecast. We couldn't really believe it, but also the stiffness numbers were uh in the forecast basically where we want them. Um and the frames came out of the mold at the forecasted weight, hit the stiffness numbers. So yeah.

SPEAKER_01 25:45

So so the the forecasted weight, the stiffness numbers, the compliance numbers, and of course, which is totally non-negotiable is the durability numbers, the testing numbers. Yeah.

SPEAKER_00 25:53

Yeah, I mean testing structural testing, I think we internally have one of the most um severe structural testing, I would say, within the industry. So obviously, I mean, we those bikes are being raised by the one of the best performance uh athletes in the world, so everything has to be safe. Obviously, that's yeah, like yeah, what you said, non-negotiable.

SPEAKER_01 26:14

Yeah. So so really just to go back to that idea of each ply, making sure it's both you know structural and providing a stiffness benefit. So really you're making sure it's the the shaping of the bike itself, not just for arrow, but for structural load. So that you don't need layers that are just providing stiffness because the shape itself, the structure, is really helping provide all that stiffness.

SPEAKER_00 26:39

Exactly, yeah. So it's uh basically a combination, I would say, of shape and layup that allows you then to um design the layup in a way that you like don't have those what we call lazy plies that you need to add that either only serve structural properties or only stiffness properties. But every single play basically is combining both uh properties and serves what I said double duty.

SPEAKER_01 27:04

Which I guess it's interesting because that all feeds into what you know you're learning using the moving like mannequin about how much the rider and the bike interaction changes what your bike's gonna look like. You you don't have to make these sort of structurally inefficient big airfoils because there's not a ton of benefit to that.

SPEAKER_03 27:20

Yeah, of course. So this bike, kind of to your point, Ben, was designed around a rider, right? I think for us that's a really key metric is bikes don't ride themselves. We are designing our bikes for the way that they're ridden, for the way that they're raced. So this bike in specific, we designed it kind of around that kind of modern breakaway arrow position, um, using our sixth generation most precise ever moving leg mannequin system. And a lot of features on this bike were informed directly by that. The most obvious of them I would say is the aero seat post. Um, your thighs are narrowing down, that accelerates the airflow. That seat post actually is both the leading edge and seeing accelerated airflow. So that is one of the most key areas on the bike for drag reduction, hence our arrow post, extremely, extremely narrow, really minimize that frontal area. And then, of course, the arrow in the front, lightweight in the back, that comes from the fact that we see much cleaner airflow, uninterrupted airflow hitting the front of that chassis. While by the time it gets to the back, adding that same amount of surface area is going to have a much lower payoff and drag reduction with the same penalty and weight.

SPEAKER_01 28:30

That's interesting. So between the rider's legs, as you hit that arrow seat post, the air is moving faster there than it's moving at the front of the bike.

SPEAKER_03 28:39

Yes. So, I mean, this is kind of Bernoulli's equation, right? If you look at the way a wind tunnel is designed, it starts really big and then necks down really narrow. As you bring that airflow down into a more constricted space, it has to speed up. Um, that's what happens between your legs as you ride. The airflow between your legs has to speed up to go between that narrower area and what's right in the middle of in between your legs is the seat post. So you have this really high speed accelerated flow hitting that region. So cutting down that thickness of the seat post as much as possible without negatively impacting compliance was a really big priority for the team.

SPEAKER_01 29:14

Wow. So, okay, so we we kind of dug into the seat post there a little bit. Let's talk about some of the other things a rider is gonna see when they they walk into the bike shop or they're watching Remco or Demi race this bike. What are the other features that are gonna stand out versus SL8 that really helped you get to that four counts or that you know that reduced drag that tarmac SL9 provides?

SPEAKER_03 29:34

Yeah, of course. And I would love for um for the team here to go into a little more detail on actually the engineering that went into those features, um, not just the arrow story behind them, right? With the head tube, we did a lot of development with the head tube, trying different widths in the wind tunnel and seeing what the payoff for every millimeter narrower that it was. Um, and then Dennis came up with an incredible way to reduce that width as much as possible because we knew the payoff was there. Our fork blades twist inwards to make sure we have that clean airflow following from the fork through the wheel through the down tube, really just cleaning everything up. Another really big distinction from SL8 is the fork crown into the down tube. The down tube dropped to meet the back of the fork crown. And we, if you just even run your finger through there, right? It's an incredibly smooth, clean transition. Really focused on cleaning up the airflow on the front of the bike, down to the fact that the fork blades were also pushed further forward, which going a little further back on the bike, we have our wind fin. That was kind of that that wind fin only matters when you don't have a bottle on the back. When you have two bottles, it doesn't add any arrow to the bike. But when you have one bottle, it actually does add arrow to the bike. So in those critical moments, when you're probably riding by yourself and seeing the full force of the wind, you get a little extra arrow boost there. Similarly, that front of the seat tube they brought a little further forward, get a little bit longer of a leading edge.

unknown 30:57

Cool.

SPEAKER_01 30:57

So let's let's talk about you. You touched on one of the engineering challenges that that trying to save, shave this drag really created. When you see bikes with really narrow front ends, you see a lot of people going to a bayonet. Right? You didn't want to go to the bayonet. Yeah. You had the ride quality, you wanted the riders to have the handling, you wanted them to be able to that they love so much. So, how do you narrow the head tube of the SL8 with internal cables without having to go to a bayonet or some other sort of novel solution?

SPEAKER_00 31:24

Yeah. Um, I think that's actually a pretty important point uh with the internal cables, because internal cables plus uh steerer, fork steer autodiameter is basically giving you uh the room that you have uh available for your um head tube width. So when we came up with the idea of reducing the head tube width by four millimeters, what we did on SL9. So overall, like if you look at the uh frontal area of the head tube, that's a minus 10% reduction in frontal area, which is pretty big for us. Um it was clear, okay, we are not able to fit brake hoses on both sides of the fork steer anymore. So there would have been two solutions basically. One would be either we reduce the outer diameter of the um fork steerer inside the head tube, or we go to a non-round uh fork steerer. Both of those options were not really an option for us, I would say, because we fork steerer is really critical for um riding characteristics, front-end stiffness. So it was clear we want to keep the fork steerer OD where we have it at uh with SL8. So the only um solution for that was basically what we did on SL9 is that we offset the fork steerer two millimeters to the non-drive side. So you have no more room for your brake hoses on the non-drive side fork uh side of the fork um steerer anymore. But your brake hose has to be routed on the drive side fork uh of the fork steer. So first time our test lab uh received fork samples and they looked at the forks, they thought something was wrong, so I got some messages out. Um so yeah, it looks quite uh quite interesting, I think, but that was definitely um a challenge.

SPEAKER_01 33:07

Yeah, just but pretty simple. Both cables just running down pretty easy, right? Yeah, running down the same side of the fork. Yeah. Steer tube amazing. Um all right, so let's take a step back. You've created this bike, same weight, two grams, 687 gram frame. Preserved that ride quality that it sounds like the athletes said, you know, whatever you do, don't mess up the way this bike handles and the way this bike feels at the pedals and feels in the saddle. So you you preserved all that. You've shaved four counts of drag, right? Which give us a quick explainer on counts in case people don't understand how a count works versus watts, which we typically hear.

SPEAKER_03 33:45

So for this bike, our arrow claim is four watts at 45 kilometers an hour. That would be just under four counts of drag. Um, a count is a unit of the actual coefficient of drag times the frontal area of the bike. In this case, at around 42 and a half kilometers an hour, one count is approximately equal to one watt. Um, but the watt is actually at the specific speed. So maybe it's four watts at 45 Ks, and then at 50 K's that's gonna be even more watts than it takes you.

SPEAKER_01 34:15

Okay, but for for regular non-math nerd guys, about four watts at 45k. Yeah. Okay, cool. Um, so you created this bike now. And to you know, your point about the simulation side, you used the equation of speed to figure out exactly what you needed to make. Once you make it, now you can kind of start to predict some things, or you can even look back and say, how would someone have performed on this bike, you know, in a previous race if they had the SL mind in their arsenal. Um any examples of that? I think there's one specifically, Alex, you have that might be a pretty compelling example of SL8 versus SL Mine?

SPEAKER_02 34:54

Yeah. When we when we had kind of uh the bike pretty much finalized, we did, let's say, uh, one of our first real-world simulations looking back at a couple different performances. One of those was from the 2024 Tour de France Fums, where Demi lost to Kazia by four seconds. So she, if we remember from that last stage, um, basically attacked with about 83K to go. She had a lot of descending, a lot of flat, and then obviously the final climb up Alp Duez. And essentially what we looked at was her complete uh fit file, obviously, like writer data there. We looked at real-world weather data as well, and Marcel analyzed all of that. And over those 83K, she essentially would have been about 14 seconds faster on SL9. So maybe sounds like not a lot for most folks, but that would go from essentially losing the Tor de France by four seconds to winning the Tor de France by 10 seconds. So huge result change there for her if she had been on the SL9. So I think that was really exciting for us to see, especially when we can um yeah, put that into a a real-world situation where it makes a huge difference for an athlete.

SPEAKER_01 36:05

Yeah, it's pretty incredible to think about that. When you know, we talk about marginal gains and you know, in shaving a gram here and shaving some drag there. And these bikes really made for racing, made for winning the biggest races in the world. This is a great example of you know what can seem a marginal gain is not so marginal at the finish line when you're trying to win a yellow jersey at the Tour de France. Absolutely. Yeah, that that's pretty phenomenal. That's really phenomenal. Um, one thing I want to just touch on, we've talked a lot about these simulations, right? And in the in-depth way you put these simulations together, the the lengths you go to to get the right data from the rider files to what's happening in the wind tunnel and the moving leg mannequin to ensure we have this incredible accuracy. I think this kind of simulation is fairly new in cycling. You know, not a lot of other products are developed using this kind of simulation. Um, why should we believe these simulation numbers? You know, it's just you're running in the computer. Who else is doing this? Is this a thing?

SPEAKER_03 36:59

Um, I'll do my best impression of our simulation engineer for you, Marcel Kaiser. He would tell you that we've been doing this since 2016, right? And it is absolutely widespread in high performance sports. If you look at any high-end motorsport, Moto GP, F1, um, sailboat racing, right? Every decision made in those disciplines comes down to these same kinds of simulations. Our simulation engineer himself came from F1, right? So this is extremely widely used in other industries. But what Marcel would also tell you is our application of it in cycling is probably unmatched in its resolution. The way that the rider has to slow down for every corner, right, to make sure it's as realistic as possible. He's had time trial simulations finished generally within plus or minus five seconds of the rider's actual time when he matches their power file. So we believe it's an industry-leading tool that we have, but the equations are easy, the physics are easy. What really is the secret sauce is the accuracy of the inputs. We have multiple treadmills within specialized that we can use for rolling resistance testing. We have the wind tunnel, we have velodrome testing, we have road testing. Um, we have these long-standing relations with all of our athletes. And just having that accuracy of the inputs really is the most important thing for us because at the end of the day, as a professor at college would love to tell any of us, garbage in, garbage out, we have to have the most accurate inputs to know that what we're saying is actually accurate on the output side.

SPEAKER_01 38:28

Yeah, that's really interesting when you think about it from like the Formula One standpoint, right? If someone got to the start line of an F1 Grand Prix and said, Well, we got the most aero car, we're gonna win. Nobody cares. Yeah, it's worthless, right? It's like everything else that goes into it, oh, we got the lightest car. Oh, our car does the most horsepower. Everyone knows that doesn't matter. It is what car does the fastest lap time.

SPEAKER_02 38:48

I think the other piece to Leo's point, though, that's kind of interesting is just going back to our relationship with the athletes. I think their fit files or um data files from a rider race are probably like one of the more secretive things that that riders and athletes have. They don't really share those unless it's like, let's say, one-on-one with their coach. But pretty much all of our athletes are willing to share any data they have with us because they understand at this point how much we utilize that to increase the accuracy of our simulations and just give them better insight to what's gonna help them perform at the highest level.

SPEAKER_01 39:22

Yeah, yeah. Amazing. So you you guys have, you know, not just the three of you, a much bigger team was a part of this. You all, you know, three played a huge role in this. You've updated this legendary iconic tarmac. It's been the road platform for specialized, like we said, that you know, no bike since it launched has won more world championships. Um, no bike since it's launched has won more races. You've really found eked out an incredible amount of performance out of a platform that was already so mature. You know, the the same weight, 6.5 kgs, out of the box builds, that same ride quality, that same stiffness, the same handling, but now with those four counts, more arrow. And you just talked about those athletes that you serve, you know, the demis and the remcos and the lottas and the Lorenas and the Pelazzaris and um Jai Henley, and in a couple days, they're going to be rolling out for the Tour de France on this bike. This thing you've been working on for the last three years. How does that feel?

SPEAKER_00 40:22

Pretty good. Pretty good, yeah. Pretty amazing feeling.

SPEAKER_01 40:25

You know, when you're watching the race, when you you know, uh all of us love to cheer, and you know, especially it's specialized, everyone's watching the races all the time. It must be a you know kind of unique feeling when it's the bike you all created.

SPEAKER_02 40:37

Yeah, I think our entire team is super proud of all the effort and work that went into it. So I feel like we all, Sarah, share a uh, let's say, sense of enjoyment when we see the athletes performing on it and also when they when they perform really well on it. So it's super fun.

SPEAKER_01 40:53

Alright, well, anything we missed in this story of updating the iconic tarmac.

SPEAKER_02 41:00

I think that covers most of it.

SPEAKER_01 41:02

All right, well, thank you guys so much for the time and taking us behind the scenes, peeling the curtain back on what it took to make the tarmac SL9. And I hope you all tune into the Tour de France to watch our athletes on the tarmac SL9 when the race starts in Barcelona and finishes all the way in Paris in uh in 21 days. And um, keep listening, subscribe and follow, and all those good things. Thank you.